188 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 31 



$55,000 to enable me to find, equip, and operate a solar-constant sta- 

 tion in the best locality of the Eastern Hemisphere. After journey- 

 ing to Algeria, Egypt, and Baluchi- 

 stan, I chose Mount Brukkaros, in 

 South West Africa, at 5,200 feet ele- 

 vation. I am sorry to report that 

 neither Table Mountain nor ISIount 

 Brukkaros equals Montezuma in 

 favorable qualifications for the solar- 

 constant work. At present, m}' col- 

 league, A. F. Moore, is testing 

 other high mountains in South West 

 Africa and the Cape Verde Islands, 

 hoping to find a site as good as 

 Montezuma. 



I have collected in Volume V of 

 the Annals of the Astrophysical 

 Observatory, recently issued, the 

 results of 10 years of solar-constant 

 work at several stations. 



MONTHLY MARCH OF SOLAR 

 VARIATION 



Monthly mean values from the 

 Avidely separated stations agree ex- 

 cellently and unite to determine the 

 principal trends of variation of the 

 solar radiation with adequate accur- 

 acy and certainty. As the probable 

 error of the general mean curve is 

 less than 0.1 per cent, we may feel 

 much confidence in solar changes as 

 small as one-fourth of 1 per cent 

 when indicated by monthly mean 

 values from three stations. The 

 maximum range of variation of the 

 general mean of the monthly mean 

 values of the solar constant since 

 1920 is 2.8 per cent, and only 1.2 per 

 cent since 1926. The variations of 

 individual days may be considerably 

 wider, so that the extreme range of 

 solar variation since 1920 appeal's to 

 be about 4 per cent. 



